3522c460e3
tweak: update transform for gemini models so that topP and topK match gemini-cli values
2025-12-25 22:46:12 -06:00
9898fbe8ef
providers: add Cloudflare AI Gateway ( #5174 )
2025-12-23 17:31:58 -06:00
526c723e62
support glm 4.7
2025-12-22 17:11:02 -05:00
87b5b34280
fix(providers.opencode): check config for api key in addition to auth ( #5906 )
2025-12-22 15:20:40 -06:00
f9be2bab3a
fix: bundle more providers to fix breaking ai sdk issue
2025-12-22 15:12:18 -06:00
8785bec29c
tweak: adjust minimax m2 topK and add minimax m2.1 topP
2025-12-21 21:17:58 -06:00
3456f4ed80
tweak: update kimi-k2 and kimi-k2-thinking to use recommended temperature values
2025-12-21 15:54:57 -06:00
d89b567b47
fix: add transform case for gemini if mcp tool has missing array items ( #5846 )
2025-12-20 11:41:52 -06:00
dacbbe3184
chore: rm dead code
2025-12-19 16:20:17 -06:00
9427f56e1a
rm interleaved thinking filter for certain kimi k2 thinking model providers that were bugged
2025-12-18 12:26:27 -06:00
d5dcc55a47
Revert "add client header"
...
This reverts commit 2fb89161c8
2025-12-18 11:21:22 -05:00
2fb89161c8
add client header
2025-12-18 10:33:25 -05:00
7d1733c752
core: fix message caching for Anthropic models to improve response consistency
2025-12-15 14:33:14 -05:00
9eefcd1b41
Provider fix, anthropic Errorhandling if empty image file is read ( #5521 )
2025-12-14 23:56:47 -06:00
4a8e8f537c
wip(desktop): progress
2025-12-14 21:38:58 -06:00
e22af25076
ci: fix test failures in CI by pre-populating models cache
...
Tests were failing in CI because the models.json cache file doesn't exist
and the data() macro fallback only works at build time, not runtime.
The preload now pre-fetches models.json and disables the background
refresh to prevent race conditions during test execution.
2025-12-14 21:23:45 -05:00
fed4776451
LLM cleanup ( #5462 )
...
Co-authored-by: GitHub Action <action@github.com >
Co-authored-by: Aiden Cline <63023139+rekram1-node@users.noreply.github.com >
2025-12-14 21:11:30 -05:00
0c7f0cfa2e
tweak: fallback to provider default for temperature
2025-12-14 11:57:12 -06:00
3d2ef28fa8
add topK function to transform, add temp defaults for glm and minimax
2025-12-13 23:27:11 -06:00
d4cfbd8219
chore: reduce duplication of field in transform
2025-12-13 18:07:22 -06:00
f9674793af
tweak: 5.1 -> 5. in transform
2025-12-12 15:17:11 -06:00
909013320b
tweak: 5.1 -> 5. reasoning effort match
2025-12-12 13:21:57 -06:00
342595e0f7
tweak: setting chat_template_args in transform
2025-12-11 16:22:10 -06:00
7b4f852f33
ignore: tmp transform exclusion
2025-12-10 21:18:19 -06:00
439aebb4e9
tweak: correct thinkingLevel
2025-12-10 21:13:10 -06:00
df55ad89ab
Add Cerebras integration header with opencode identifier ( #5354 )
2025-12-10 19:36:58 -06:00
ffc4d53923
add models.dev family
2025-12-10 14:30:28 -05:00
18d3c054a3
more interleaved thinking fixes ( #5334 )
2025-12-10 13:29:26 -06:00
59c5da9b6c
tweak: set gemini thinkingLevel to high by default
2025-12-10 13:09:27 -06:00
117de64f39
fix: upgrade OpenRouter sdk pkg ( #5302 )
...
Co-authored-by: Aiden Cline <aidenpcline@gmail.com >
2025-12-10 13:02:26 -06:00
df64612d54
better interleaved thinking support ( #5298 )
2025-12-09 16:32:12 -06:00
f8bd4ff705
core: refactor providerOptions function to accept Provider.Model for cleaner API
2025-12-09 01:25:36 -06:00
52bb43eebd
fix: SAP AI Core Vercel AI SDK v2 support ( #5180 )
...
Signed-off-by: Jérôme Benoit <jerome.benoit@sap.com >
Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org >
2025-12-08 11:27:47 -06:00
a3bb4a3c85
pdf support in read tool ( #5222 )
...
Co-authored-by: ammi1378 <ammi1378@users.noreply.github.com >
2025-12-07 21:54:00 -06:00
c30b1130ee
fix: provider url merging logic
2025-12-07 13:01:05 -06:00
1b05d5dd8e
tui: prevent deprecated models from appearing in model picker
2025-12-06 18:18:45 -05:00
52db0f23a6
fix : #5064 ProviderInitError github-copilot-enterprise ( #5123 )
2025-12-05 09:53:32 -06:00
bef4fdfc4b
fix: add getModel to SAP AI Core provider for correct SDK initialization ( #5086 )
...
Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org >
2025-12-04 22:43:22 -06:00
668d5a76d5
core: ensure model npm package falls back to dev models config when not explicitly defined
2025-12-04 15:39:52 -05:00
b9c1f10016
feat: Add SAP AI Core provider support ( #5023 )
...
Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org >
2025-12-04 14:07:23 -06:00
10cc15aabe
fix anthropic api key error
2025-12-03 23:18:24 -05:00
46790e57e9
feat: Enhance DeepSeek reasoning content handling ( #4975 )
...
Co-authored-by: Aiden Cline <aidenpcline@gmail.com >
2025-12-03 21:31:36 -06:00
598d63db63
fix: dax typo
2025-12-03 20:39:11 -06:00
6d3fc63658
core: refactor provider and model system ( #5033 )
...
Co-authored-by: opencode-agent[bot] <opencode-agent[bot]@users.noreply.github.com>
Co-authored-by: thdxr <thdxr@users.noreply.github.com >
2025-12-03 21:09:03 -05:00
ee4437ff32
core: add provider test coverage for upcoming refactor
...
Add comprehensive test suite for Provider module to ensure safe
refactoring of provider internals. Tests cover:
- Provider loading from env vars and config
- Provider filtering (disabled_providers, enabled_providers)
- Model whitelist/blacklist
- Model aliasing and custom providers
- getModel, getProvider, closest, defaultModel functions
Also adds Env module for instance-scoped environment variable access,
enabling isolated test environments without global state pollution.
2025-12-03 18:30:42 -05:00
b457923970
core: fix GitHub Copilot Enterprise authentication failing with sdk.chat undefined error
2025-12-02 23:37:10 -06:00
0331931f56
fix: sanitize more invalid schema cases for gemini models
2025-12-01 16:19:10 -06:00
f1bb5870ce
fix: copilot responses bug w/ gpt 5 mini
2025-12-01 14:36:49 -06:00
027d43b5ea
fix case where opencode wasn't retrying
2025-12-01 13:02:33 -06:00
b1b82977ec
tweak: better err msgs
2025-12-01 01:33:32 -06:00
Aiden Cline